The Transform Technology Summits start October 13th with Low-Code/No Code: Enabling Enterprise Agility. Register now!
Clusterone today announced the close of a $2 million funding round to help data scientists and businesses automate and optimize infrastructure management necessary to deploy AI systems and carry out machine learning workflows.
The platform can operate with both on-premise servers and popular public cloud computing platforms like AWS, Azure, and Google Cloud Platform.
Clusterone is made particularly to take infrastructure management responsibilities out of the hands of IT teams so they can focus on making AI models instead of basic DevOps tasks.
Using the latest software and up-to-date containers and intelligent tracking of resource utilization and things like spot instances can help businesses significantly reduce their cloud computing costs, CEO and cofounder Mohsen Hejrati told VentureBeat in a phone interview.
The platform acts as middleware and is especially geared for businesses that have yet to determine if their AI should operate from on-premise servers or in the cloud, he said.
“Our customers, they don’t know what’s the right thing to commit to at this point,” Hejrati said. “Nobody knows in the market, so having that kind of freedom is very important and the way we think about it is that we want to build an operating system. So we give the customers the flexibility to choose the infrastructure and the flexibility to choose any application they want to run on that infrastructure. So we are that middle layer where, you know, we figure out all the heavy lifting of the DevOps piece and then provide kind of a unified user experience.”
The Clusterone platform allows users to utilize anywhere from a single node to distributed learning across dozens of processing units to train and deploy AI models.
In addition to DevOps management, Clusterone wants to be a single platform for machine learning teams within companies to collaborate or work on selecting the best AI models and data collection approaches possible.
Clusterone is one of a short list of companies to participate in the Allen Institute for AI Incubator, which branched off from the AI research institute in January 2017. In addition to working with early-stage companies, the incubator works with large corporations, and operates a residency program for talented technologists.
Hiring highly paid talent and having them manage your server infrastructure is a waste of their time and talent, sort of like making LeBron James mop the court before a basketball game, Allen Institute incubator director Jacob Colker told VentureBeat in a phone interview.
“These companies are hiring these superstar data scientists. They’re highly paid folks, and they show up for work, and they want to work on solving a business problem, running experiments, and tracking and executing those experiments, but there is no infrastructure to do that currently in the enterprise world,” Colker said. “So what ends up happening is you have these highly paid, highly talented people doing kind of these wrote DevOps tasks just to be able to do their job.”
The Allen AI Institute was founded four years ago by Microsoft cofounder Paul Allen and Oren Etzioni to explore things like bringing common-sense intelligence to deep learning. Since then, the institute has published about 175 research papers.
Clusterone was founded in October 2016 and is based in Seattle. The company has 18 employees and offices in San Francisco, Toronto, Canada, and Gdansk, Poland.
VentureBeatVentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative technology and transact. Our site delivers essential information on data technologies and strategies to guide you as you lead your organizations. We invite you to become a member of our community, to access:
- up-to-date information on the subjects of interest to you
- our newsletters
- gated thought-leader content and discounted access to our prized events, such as Transform 2021: Learn More
- networking features, and more